Harnessing the Predictive Power of High-Performance Computing

IBM and the University of Michigan (UM) are collaborating to develop “data-centric” supercomputing systems designed to increase the pace of scientific discovery in fields as diverse as aircraft and rocket engine design, cardiovascular disease treatment, materials physics, climate modeling and cosmology.

The system is designed to enable high-performance computing applications "for physics to interact, in real time, with big data" to improve scientists’ ability to make quantitative predictions, the organizations say. IBM’s systems use a GPU-accelerated, data-centric approach, integrating massive datasets with high-performance computing to develop new predictive simulation techniques to expand scientific knowledge.

The organizations have designed a computing resource called ConFlux to enable high-performance computing clusters to communicate directly and at interactive speeds with data-intensive operations. Hosted at UM, the project establishes a hardware and software ecosystem to enable large-scale data-driven modeling of complex physical problems, such as the performance of an aircraft engine, which consists of trillions of molecular interactions.

Cognitive techniques will be used to simulate turbulence around aircraft and rocket engines. Image credit: Pixabay.

“There is a pressing need for data-driven predictive modeling to help re-envision traditional computing models in our pursuit to bring forth groundbreaking research,” says Karthik Duraisamy, assistant professor in the UM Department of Aerospace Engineering. “ConFlux allows us to bring together large-scale scientific computing and machine learning for the first time to accomplish research that was previously impossible.”

One of the first projects to be undertaken with the advanced supercomputing system involves working with NASA to use cognitive techniques to simulate turbulence around aircraft and rocket engines. Data from wind tunnel experiments and simulations will be combined to build computing models used to predict the aerodynamics around new configurations of an aircraft wing or engine.

With ConFlux, turbulence can more accurately be modeled and studied, helping to speed development of more efficient airplane designs. It will also improve weather forecasting, climate science and other fields that involve the flow of liquids or gases, the organizations say.

UM is also studying cardiovascular disease for the National Institutes of Health. By combining noninvasive imaging such as results from MRI and CT scans with a physical model of blood flow, UM hopes to help doctors estimate artery stiffness within an hour of a scan, serving as an early predictor of diseases such as hypertension.

Studies are also planned to better understand climate science, including how clouds interact with atmospheric circulation, the origins of the universe and predictions of the behavior of biologically inspired materials.